Storing game high scores for legacy games on TNFS servers.

[tschak909]

Zombies REALLY doesn't like the 130XE I have out on the bench at the moment...

[tschak909]

I pulled into Omnivore for a look, and it looks like it's checking TRAMSZ. ugh.  Let's just patch right over that silliness...

[tschak909]

New #Atari8bit High Score Enabled game for #FujiNet: Tutti Frutti from Adventure Int'l!

TNFS: apps.irata.online/Games/High Score Enabled/Tutti Frutti (BASIC).atr

Scores: https://scores.irata.online/tuttifrutti.html

Enjoy! #retrogaming #retrocomputing 

[tschak909]

Added Gorf and Wizard of Wor from Roklan, as they also save their scores.

 

Another potential game is World Karate Championship. It has a high score table, but it isn't persisted.

I found the table in display memory, needs to be found on disk.

[+DjayBee]

Some more:

 

672-673 Lode Runner

672-673 Championship Lode Runner

137 Baja Buggies

26-29 Silent Service

90 Serpentine

718-719 Summer Games

25-26, 408 Flak (408 is no high-score sector but must be writable for the game to run)

 

Lode Runner (1984)(Broderbund Software)(US)(Side B)[!][cr CSS].atr Baja Buggies (1982)(Gamestar)(US)[!][OS-B][cr CSS].atr Championship Lode Runner (1985)(Broderbund Software)(US)[m][cr CSS].atr Silent Service (1987)(MicroProse Software)(US)(Side B)[m][a][cr CSS].atr Serpentine (1982)(Broderbund Software)(US)[!][cr CSS].atr Summer Games (1984)(Epyx - Rushware)(DE-US)(Side A)[!][cr CSS].atr Summer Games (1984)(Epyx - Rushware)(DE-US)(Side B)[!].atr Flak (1984)(Funsoft)(US)[turn off write-protect][f][cr CSS].atr

[+DjayBee]

9-10 Snokie

Snokie (1983)(Funsoft)(US)[cr CSS].atr

[Tron04]

I played a bit Jawbreaker II and could enter highscore. There were already some other entries after I have booted the game.

After playing and entering the highscore, it seems that the highscores all reset to 00000.

Because when I played the second time I could enter my name again, but was suddenly alone at first position in the highscore list.

I guess this game needs some fixing. I mounted the game of course readonly from apps.irata.online

[+DjayBee]

15 minutes ago, Tron04 said:

I played a bit Jawbreaker II and could enter highscore. There were already some other entries after I have booted the game.

After playing and entering the highscore, it seems that the highscores all reset to 00000.

Because when I played the second time I could enter my name again, but was suddenly alone at first position in the highscore list.

I guess this game needs some fixing. I mounted the game of course readonly from apps.irata.online

Do you have the latest firmware for Fujinet because earlier versions do not support high-score saving?

[chad5200]

Would it be possible to hack some of these high score enabled games to make it so difficulty cannot be changed?  To truly level the playing field.

 

For example, Pac-Man lets you adjust starting difficulty.  The problem is that most people will play this game with normal difficulty (start game on the Cherry level where the Cherry is worth 100 points).

 

But you can adjust the starting level all the way up to Key which is worth 5000 points per Key eaten.

 

This completely unbalances the game as really the only way to efficiently compete on the public high score table is to know that you should start with Key level difficulty.

 

Or if there was a way to note which difficulty level the player started with in the high score table.

 

I am using Pac-Man as an example, but there might be others as well.  Pac-Man competition would be much more interesting if the scores were wiped and everyone was forced to start on the same Cherry level.

 

 

[Tron04]

12 hours ago, DjayBee said:

Do you have the latest firmware for Fujinet because earlier versions do not support high-score saving?

Yes, I reverified the Firmware Version in the config screen and it is: 0.5.7eb55adb

[+DjayBee]

24 minutes ago, Tron04 said:

Yes, I reverified the Firmware Version in the config screen and it is: 0.5.7eb55adb

Meanwhile somebody else on Discord wrote about the same problem. 

Let's hope @tschak909 will have a look at his scraper tool for Jawbreaker II. 

[tschak909]

I have uploaded a patch for miner 2049er. It's finished, but needs more debugging. Maybe somebody can help.

https://github.com/FujiNetWIFI/fujinet-high-scores/tree/main/atari-game-ports/miner-2049er

 

-Thom

[tschak909]

1 hour ago, DjayBee said:

Meanwhile somebody else on Discord wrote about the same problem. 

Let's hope @tschak909 will have a look at his scraper tool for Jawbreaker II. 

I just posted a new score to JawBreaker II. hm

 

 

 

 

[+DjayBee]

Found yesterday at ABBUC's general meeting:

 

719-720 Knock Out

Knock Out (1985)(Kingsoft)(DE)[cr CSS].atr

[Wrathchild]

Just read this thread for the first time.

One of the issues is typically that the h/s table is only read / setup when the game is loaded.

I would suggest that an approach to patching any game would be to identify at which point the game begins scanning to see where your score will fall.

A hook should be inserted to then reload the h/s table from the ATR, hence this triggering the loading of any scores uploaded in the meantime.

Upon return your score is compared against the latest and so if you were successful this will be entered (or you asked to enter name/initials etc).

Then the write back to the ATR can occur and the master h/s table made available for other players.

This still leaves a window where another player (P2) completes a game around the same time as you (P1) but takes longer to enter their initials.

In this case P1 & P2 both have loaded HSa. P1 got top spot and enters their initials and saves as HSb. P2 completes their initials and saves, HSc, but P2's HSc doesn't contain P1's score and so that has been lost.

 

Personally I'd prefer to see the atarionline.pl Hi-Score Café utilised and instead oatch titles to post scores directly there.

This is fine with 8K,16K, 32K titles where there is plenty of free RAM to add the necessary code and patch in the hook to it.

But granted that for larger games this becomes a bit harder as that space might be harder to come by.

However that doesn't exclude employing an overlay and loading that in to do the job and then restoring the game code.

[tschak909]

I appreciate your insights.

 

I did the High Score Enabled feature as a way to bring existing games into the fold for playing existing arcade games together.

 

As such, Making changes to existing games does mean that some shortcuts had to be made, one of them was understanding that there were a number of race conditions that can occur. 

 

This particular method is a complement to atarionline.pl's method that does not require an external server, and can be deployed quickly at local parties.

 

If you want to take a crack at improving on what I've done, go ahead, the code is available for virtually every part of it.

https://github.com/FujiNetWIFI/fujinet-high-scores

 

-Thom

[+DjayBee]

8 hours ago, Wrathchild said:

Personally I'd prefer to see the atarionline.pl Hi-Score Café utilised and instead oatch titles to post scores directly there.

This certainly is the more sophisticated solution but also magnitudes more complex to implement into existing programs.

For Thom's solution you only need to add code which reads and writes one or more sectors on disk. The remaining stuff is handled by Fujinet and the backend.

 

Also, due to the huge amount of people playing these games, the risk that two of them are high-scoring at the same time is probably quite close to NIL.

[+DjayBee]

@tschak909 a very nice Roc'N'Rope clone:

 

178 Tales of Dragons and Cavemen (92 90 8D 89 89 means 97400 -> byte-$89=digit)

 

and

720 Time Bandit

 

Tales of Dragons & Cavemen (1986)(AMC-Verlag)(DE)(FW)[req 64K][cr CSS].atr Time Bandit (1984)(Atari)(DE)[cr CSS].atr

[tschak909]

4 hours ago, DjayBee said:

@tschak909 a very nice Roc'N'Rope clone:

 

178 Tales of Dragons and Cavemen (92 90 8D 89 89 means 97400 -> byte-$89=digit)

 

and

720 Time Bandit

 

Tales of Dragons & Cavemen (1986)(AMC-Verlag)(DE)(FW)[req 64K][cr CSS].atr 90.02 kB · 0 downloads Time Bandit (1984)(Atari)(DE)[cr CSS].atr 90.02 kB · 0 downloads

ooo! awesome, will put these up!

 

-Thom

[+DjayBee]

200-204 Herbert

start game, do nothing and die to see the score table

boot side B

side A with [m] contains a modified HS table

 

550-551 and perhaps 368 Ace of Aces

start a mission and push the plane to the ground a.k.a. die and then push the trigger, to see the score table

Both images have a lot of differences, so do not compare them. But they share the main code and the HS table.

The US Gold release contains a modified HS table and a modified sector 368 of which I don't know if it is HS-related or a plain dumping error.

 

If a good player of Ace of Aces reads this: Score high on the attached Accolade image and post the modified image - thanks.
 

Herbert (1988)(AMC-Verlag)(PAL)(DE)(FW)(Side A).atr Herbert (1988)(AMC-Verlag)(PAL)(DE)(FW)(Side A)[m].atr Herbert (1988)(AMC-Verlag)(PAL)(DE)(FW)(Side B)[!][cr CSS].atr Ace of Aces (1987)(Accolade)(US)[!][req 64K][cr CSS].atr Ace of Aces (1987)(US Gold)(GB)[req 64K][m].atr

Edited December 3, 2022 by DjayBee

[+DjayBee]

324 Glaggs It! (a KLAX clone)

The first 20 bytes contain the 10 top scores followed by the 10 players' names starting from the 21st byte.

The scores are divided by 10 and the result is stored in binary format (31040 becomes 3104=$C20 and thus 20 0C on disk).

Glaggs It! (1990)(Raetz)(DE).atr

[tschak909]

Awesome, thanks, I'll get these new ones up soon!

 

-Thom

 

[+DjayBee]

624-631 Beach Head

To see the HS table boot and do nothing until the end of the self-running demos.

Beach Head (1984)(Access Software)(US)[!][cr CSS].zip

[+DjayBee]

361, 261 Poker Solitaire 2023

It writes to a file PSHI.DAT in plain text. You might add sector 262 as well, in case the file gets bigger than one sector.

See http://www.atarimania.com/game-atari-400-800-xl-xe-poker-solitaire-2023_41591.html

Poker Solitaire 2023.atr

[tschak909]

@DjayBee do you have a patched copy of Drol? It seems to have a high score table.  and it does write to disk.

-Thom